LLM inference in C/C++
C++ ggml
vulkan: work around NV bug with argsort_large.comp (#28975)
Jeff Bolz4bc272fd729b
diff --git a/ggml/src/ggml-vulkan/vulkan-shaders/argsort_large.comp b/ggml/src/ggml-vulkan/vulkan-shaders/argsort_large.compindex b2df441374..f6a29be29a 100644--- a/ggml/src/ggml-vulkan/vulkan-shaders/argsort_large.comp+++ b/ggml/src/ggml-vulkan/vulkan-shaders/argsort_large.comp@@ -27,6 +27,8 @@ layout (push_constant) uniform parameter {uint inner_end;} p;+shared int s;+void argsort(bool needs_bounds_check, const uint row) {// bitonic sortint col = int(gl_GlobalInvocationID.x);@@ -35,6 +37,12 @@ void argsort(bool needs_bounds_check, const uint row) {const uint row_offset = row * p.ncols;uint idx_offset = row * p.ncols_padded;+ // workaround for NV driver/compiler bug - dummy use of shared memory+ if (gl_LocalInvocationIndex == 0) {+ s = 0;+ col += s;+ }+bool need_barrier = false;// initialize indices